Coaches, I'm happy to share the website link: https://www.thenextguess.com/

Please give it a try and see how it works for you.

A few additional comments:

1. I think that you will find the team rankings mostly mirror GUESS. If you see something that looks completely out of touch please let me know.
2. I've made a cursory attempt to make the site work reasonably ok on mobile phone screen sizes, but haven't tested extensively. You will have to zoom and/or pan/scroll to view all data. Let me know if something seems completely broken.
3. I took a shot at Team level offensive and defensive grades for run and pass. You will see these on each team's individual page. My "enhancement" is to break down the grades by formation. At the simplest level, I fill out the roster by taking the top players at each position for each formation based on the number of positions required. I try to pick players based on the positional role ratings that make sense for run vs. pass. I then calculate an overall team rating for that formation which is then normalized against all other teams in the division. That normalized score is used for the grade. Similar to GUESS, it only counts the "starters". It does not measure the value of any backup players. Something I may choose to add later. It also doesn't factor in formation ratings as I wasn't sure how to best factor that in. I can't claim my approach to team grades is any better than what GUESS currently has. It might be way worse. Will be interested to hear your feedback.

Please continue to thank Bob for the amazing work he has done with GUESS over the years. I hope The Next Guess can continue carrying the torch.

looks pretty sweet

Outstanding! A tribute to Bob and now to you. I appreciate it. Some bugs I am sure to solve moving forward but a grand start. Cheers!

Amazing! Thank you!

Good stuff.

I really like the introduction of relative values for players, that is key.

In the original GUESS, the top player lists seemed to always be dominated by OLs, DLs, (and sometimes RBs) because of the way that the formulas were calculated - some positions have naturally higher total scores than others. The result was that based on those raw totals you would often have 20-30 (or more) OLs listed before the best QB made an appearance.

With relative values, the lists look a lot more varied and balanced.

The new formation ratings feature I am less sure about...what is the methodology for those ratings?

Great work ebzimny, like the modernized look and feel of the site.

Well done ebzimny!!!

Thank you!

great job, appreciate it

Posted by narcotico on 8/15/2023 3:56:00 PM (view original):
Good stuff.

I really like the introduction of relative values for players, that is key.

In the original GUESS, the top player lists seemed to always be dominated by OLs, DLs, (and sometimes RBs) because of the way that the formulas were calculated - some positions have naturally higher total scores than others. The result was that based on those raw totals you would often have 20-30 (or more) OLs listed before the best QB made an appearance.

With relative values, the lists look a lot more varied and balanced.

The new formation ratings feature I am less sure about...what is the methodology for those ratings?

The formation ratings are something that I really like, in concept. That is a tough nut to crack for sure, but if you share your logic here, I'm sure you will get more debate and discussion than you could possibly wish for!!! Or, you can throw it out to a few folks to give private feedback. Either way, I have some thoughts, but unsure what your formula/approach is.

If you use the new site please send a donation Eb's way. He took the time to build us awesome tools that enhance the game. Show a little love! I'm sure even $5 to $10 would be appreciated!

GUESS has relative values today, but GUESS isn't using relative values to rank all positions together, which is why we would see certain positions dominate rankings.

For letter grades by formation, it is my first attempt at it. I'm not married to it. It is not simple to get this right. Might not even be possible to get it right. Some background...

I fill out a formation by picking top rated players for each position role rating. So one thing you will notice is that a team who doesn't have enough players to fill a certain position will have a lower letter grade, likely a C/D/F grade (e.g. if you only have 2 WR's = poor letter grade for Shotgun).

For the actual calculation, my concept is similar to how GUESS does it but not identical. Each formation has a total of 22 "points" to distribute across the 11 positions. I won't go into the full detail for every formation, but here is the general concept using Pro Set as an example.

Pro Set Running - RB's would be weighted the highest
QB - take best QB Run role rating
RB1 - take the best overall RB role rating
RB2 - take the next best RB with highest FB role rating
WR1 - take best WR rating
WR2 - take next best WR rating
TE - take the best TE Blocking role rating
OL - take the top 5 OL Run role rating

SUM the weighted ratings and then calculate relative value against all other schools for Pro Set Running.

Pro Set Passing - QB would be weighted the highest
QB - take the best QB passing role rating
RB1 - take the best RB possession role rating
RB2 - take the next best RB possession role rating
WR1 - take the best WR possession role rating
WR2 - take the next best WR possession role rating
TE - take the best TE possession role rating
OL - take the top 5 OL Pass role ratings

SUM the weighted ratings and then calculate relative value against all other schools for Pro Set Passing.

Thank you,

This is really good and honors all the work Bob has done over the years.

Kudos and will be donating to you Ebz

Posted by ebzimny01 on 8/17/2023 10:02:00 AM (view original):
GUESS has relative values today, but GUESS isn't using relative values to rank all positions together, which is why we would see certain positions dominate rankings.

For letter grades by formation, it is my first attempt at it. I'm not married to it. It is not simple to get this right. Might not even be possible to get it right. Some background...

I fill out a formation by picking top rated players for each position role rating. So one thing you will notice is that a team who doesn't have enough players to fill a certain position will have a lower letter grade, likely a C/D/F grade (e.g. if you only have 2 WR's = poor letter grade for Shotgun).

For the actual calculation, my concept is similar to how GUESS does it but not identical. Each formation has a total of 22 "points" to distribute across the 11 positions. I won't go into the full detail for every formation, but here is the general concept using Pro Set as an example.

Pro Set Running - RB's would be weighted the highest
QB - take best QB Run role rating
RB1 - take the best overall RB role rating
RB2 - take the next best RB with highest FB role rating
WR1 - take best WR rating
WR2 - take next best WR rating
TE - take the best TE Blocking role rating
OL - take the top 5 OL Run role rating

SUM the weighted ratings and then calculate relative value against all other schools for Pro Set Running.

Pro Set Passing - QB would be weighted the highest
QB - take the best QB passing role rating
RB1 - take the best RB possession role rating
RB2 - take the next best RB possession role rating
WR1 - take the best WR possession role rating
WR2 - take the next best WR possession role rating
TE - take the best TE possession role rating
OL - take the top 5 OL Pass role ratings

SUM the weighted ratings and then calculate relative value against all other schools for Pro Set Passing.

I'm not entirely sure what the best practical applications for it are(i.e. teams probably won't switch formations based on one season's personel groupings), but your formation rating system looks logical and well-conceived.

Good stuff.

Bryant wraps up their recruiting period tonight. Really want to compare the Next Guess versus the Old Guess (since that one will still be around for another 6 weeks or so). I think a comparison is highlight some of the differences and enhancements.